Fix output dtype of Ops in tensor.nnet

theano/tensor/nnet/sigm.py

@@ -31,6 +31,11 @@ class ScalarSigmoid(scalar.UnaryScalarOp):
             return 0.0
         if x > 30.0:
             return 1.0
+        # If x is an int8 or uint8, numpy.exp will compute the result in
+        # half-precision (float16), where we want float32.
+        x_dtype = str(getattr(x, 'dtype', ''))
+        if x_dtype in ('int8', 'uint8'):
+            return 1.0 / (1.0 + numpy.exp(-x, sig='f'))
         return 1.0 / (1.0 + numpy.exp(-x))
 
     def impl(self, x):
@@ -268,8 +273,11 @@ def hard_sigmoid(x):
     Removing the slope and shift does not make it faster.
 
     """
-    slope = 0.2
-    shift = 0.5
+    # Use the same dtype as determined by "upgrade_to_float",
+    # and perform computation in that dtype.
+    out_dtype = scalar.upgrade_to_float(scalar.Scalar(dtype=x.dtype))[0].dtype
+    slope = tensor.constant(0.2, dtype=out_dtype)
+    shift = tensor.constant(0.5, dtype=out_dtype)
     x = (x * slope) + shift
     x = tensor.clip(x, 0, 1)
     return x
@@ -300,6 +308,11 @@ class ScalarSoftplus(scalar.UnaryScalarOp):
             return 0.0
         if x > 30.0:
             return x
+        # If x is an int8 or uint8, numpy.exp will compute the result in
+        # half-precision (float16), where we want float32.
+        x_dtype = str(getattr(x, 'dtype', ''))
+        if x_dtype in ('int8', 'uint8'):
+            return numpy.log1p(numpy.exp(x, sig='f'))
         return numpy.log1p(numpy.exp(x))
 
     def impl(self, x):